An effective anti-HIV vaginal microbicide is likely to prevent millions of new HIV infections in women. Focusing on compounds with known viral specific activity, combinatorial approaches to increase the breadth of protection and intensifying preclinical testing will greatly facilitate in meeting the challenge of developing a broadly effective microbicide. While the macaque model has been valuable for small scale microbicide testing, exploitation of a small animal model of HIV infection will greatly facilitate large scale rapid evaluation of potential microbicide compounds. In this regard, the newly developed humanized mouse models that harbor human HIV target cells show great promise. We and others have recently succeeded in achieving HIV-1 mucosal transmission via both vaginal and rectal routes in these models. This exciting development together with the discovery of new anti-retroviral compounds with novel mechanisms of action and identification of potent broadly neutralizing anti-HIV antibodies (bNAb) provides us with a unique opportunity to test new approaches for prevention of HIV sexual transmission. Our major goal in these studies employing HIV itself as a challenge virus in humanized mice is to identify new classes of molecules as potential HIV microbicides for further development. We recently obtained promising preliminary data on maraviroc, raltegravir and VRC01 bNAb as potential microbicides which form a basis for further studies proposed here. Our specific aims are to 1. Develop the integrase inhibitor raltegravir and CCR5 inhibitor maraviroc as vaginal microbicides. 2. Evaluate broadly neutralizing anti-HIV antibodies as potential HIV microbicides. 3. Evaluate important factors such as estrus period, cell-associated virus, field strains and drug resistant mutants for the success of microbicides in the field. 4. Evaluate a combinatorial microbicide strategy using select anti-HIV agents with different modes of action. 5. Evaluate the safety and potential toxicities of promising microbicide candidates in vivo.